This article is a Plant Cell Advance Online Publication. The date of its first appearance online is the official date of publication. The article has been
edited and the authors have corrected proofs, but minor changes could be made before the final version is published. Posting this version online
reduces the time to publication by several weeks.
Structure of the Arabidopsis Glucan Phosphatase
LIKE SEX FOUR2 Reveals a Unique Mechanism
for Starch Dephosphorylation
W
David A. Meekins,a,1 Hou-Fu Guo,a,1 Satrio Husodo,a Bradley C. Paasch,a Travis M. Bridges,a Diana Santelia,b
Oliver Kötting,c Craig W. Vander Kooi,a and Matthew S. Gentrya,2
a Department
of Molecular and Cellular Biochemistry and Center for Structural Biology, University of Kentucky, Lexington,
Kentucky 40535-0509
b Institute of Plant Biology, University of Zürich, 8092 Zurich, Switzerland.
c Institute for Agricultural Sciences, ETH Zürich, 8092 Zurich, Switzerland
Starch is a water-insoluble, Glc-based biopolymer that is used for energy storage and is synthesized and degraded in a diurnal
manner in plant leaves. Reversible phosphorylation is the only known natural starch modification and is required for starch
degradation in planta. Critical to starch energy release is the activity of glucan phosphatases; however, the structural basis of
dephosphorylation by glucan phosphatases is unknown. Here, we describe the structure of the Arabidopsis thaliana starch
glucan phosphatase LIKE SEX FOUR2 (LSF2) both with and without phospho-glucan product bound at 2.3Å and 1.65Å,
respectively. LSF2 binds maltohexaose-phosphate using an aromatic channel within an extended phosphatase active site and
positions maltohexaose in a C3-specific orientation, which we show is critical for the specific glucan phosphatase activity of
LSF2 toward native Arabidopsis starch. However, unlike other starch binding enzymes, LSF2 does not possess a carbohydrate
binding module domain. Instead we identify two additional glucan binding sites located within the core LSF2 phosphatase
domain. This structure is the first of a glucan-bound glucan phosphatase and provides new insights into the molecular basis
of this agriculturally and industrially relevant enzyme family as well as the unique mechanism of LSF2 catalysis, substrate
specificity, and interaction with starch granules.
INTRODUCTION
As the major energy cache for plants and algae, starch is a central
component of human and animal diets and a key constituent in
many manufacturing processes. Starch is composed of two Glc
polymers: amylopectin (75 to 90%) and amylose (10 to 25%) (Tester
et al., 2004; Streb and Zeeman, 2012). Structurally, amylose is
a linear molecule composed of a-1,4-glycosidic–linked chains with
few branches, while amylopectin is formed from a-1,4-glycosidic–
linked chains with a-1,6-glycosidic branches, similar to glycogen
in animal tissues. However, compared with glycogen, the glucan
chains composing amylopectin are longer with fewer branches
that are organized into clusters at regular intervals (Gallant et al.,
1997; Buléon et al., 1998; Roach, 2002). The biophysical properties of amylopectin result in the formation of double helices by
adjacent glucan chains that interact to form crystalline lamellae,
significantly contributing to the water insolubility of starch (Buléon
et al., 1998; Streb and Zeeman, 2012).
Starch is synthesized and degraded in a diurnal manner in
plant leaves via the concerted activity of starch synthases,
branching enzymes, isoamylases, and amylases (Keeling and
1 These
authors contributed equally to this work.
correspondence to [email protected].
The authors responsible for distribution of materials integral to the
findings presented in this article in accordance with the policy described
in the Instructions for Authors (www.plantcell.org) are Matthew S. Gentry
([email protected]) and Craig W. Vander Kooi (craig.vanderkooi@
uky.edu).
W
Online version contains Web-only data.
www.plantcell.org/cgi/doi/10.1105/tpc.113.112706
2 Address
Myers, 2010; Streb and Zeeman, 2012). Water insolubility is an
essential feature of starch that underlies the ability of starch granules to function in energy storage by regulating the access of starch
hydrolyzing enzymes (e.g., amylases) (Caspar et al., 1991; Yu et al.,
2001; Blennow and Engelsen, 2010). Reversible starch phosphorylation, via glucan dikinases and phosphatases, is necessary to
solubilize glucans in the outer layer of the starch granule and permit
their catabolism during nonphotosynthetic periods (Fettke et al.,
2009; Blennow and Engelsen, 2010; Kötting et al., 2010; Streb and
Zeeman, 2012). Thus, starch degradation is based on a cyclical
enzymatic process involving glucan dikinases, amylases, and glucan phosphatases.
In plants, there are two glucan dikinases that phosphorylate
starch. a-Glucan water dikinase (GWD) phosphorylates the C6position of Glc moieties on the starch granule surface, and this
event triggers C3-phosphorylation by phosphoglucan water dikinase (PWD) (Ritte et al., 2002; Baunsgaard et al., 2005; Kötting
et al., 2005; Ritte et al., 2006). Biophysical studies suggest that
C3-phosphorylation imposes steric effects that result in amylopectin helix unwinding and local solubilization of surface glucans,
thus permitting amylolytic degradation by b-amylases (Edner
et al., 2007; Hansen et al., 2009; Blennow and Engelsen, 2010).
However, b-amylase activity is inhibited when a phosphate group
is reached; therefore, glucan phosphatases must release phosphate from starch to allow processive starch breakdown (Takeda,
1981; Kötting et al., 2009). Plants contain two glucan phosphatases that dephosphorylate starch and allow further degradation
by b-amylase (Niittylä et al., 2006; Gentry et al., 2007; Kötting
et al., 2009; Santelia et al., 2011).
The Plant Cell Preview, www.aspb.org ã 2013 American Society of Plant Biologists. All rights reserved.
1 of 13
2 of 13
The Plant Cell
Glucan phosphatases are members of the protein tyrosine
phosphatase (PTP) superfamily characterized by a conserved Cx5R
catalytic motif (Yuvaniyama et al., 1996; Gentry et al., 2007; Tonks,
2013). The PTPs include a heterogeneous group of phosphatases
called the dual specificity phosphatases (DSPs) that dephosphorylate phospho-Ser, -Thr, and -Tyr of proteinaceous substrates
as well as more diverse substrates such as lipids, nucleic acids,
and glucans (Alonso et al., 2003; Tonks, 2006; Moorhead et al.,
2009). The glucan phosphatase STARCH EXCESS4 (SEX4) has
been shown to preferentially dephosphorylate the C6-position, and
LIKE SEX FOUR2 (LSF2) specifically dephosphorylates the C3position of Glc moieties (Hejazi et al., 2010; Santelia et al., 2011).
However, the structural basis for specific glucan phosphatase activity and position specificity has not been determined. LSF2 and
SEX4 are DSPs that are conserved in Archaeplastida/Plantae genomes from land plants to single-cell green algae (Gentry et al.,
2007; Gentry and Pace, 2009; Santelia et al., 2011). Arabidopsis
thaliana lacking SEX4 activity have larger starch granules, more leaf
starch, and altered patterns of starch phosphorylation, a phenotype
further exacerbated upon the simultaneous loss of LSF2 activity
(Zeeman et al., 2002; Niittylä et al., 2006; Kötting et al., 2009;
Santelia et al., 2011). Furthermore, LSF2 and SEX4 are functionally
related to the glycogen phosphatase laforin that is found in all
vertebrates and a subset of unicellular protozoa (Worby et al., 2006;
Gentry et al., 2007, 2009; Tagliabracci et al., 2007). Mutations in the
gene that encodes laforin in humans cause the accumulation of
insoluble carbohydrates leading to the fatal epileptic disorder Lafora’s disease (Minassian et al., 1998; Serratosa et al., 1999; Gentry
et al., 2009). These findings demonstrate that glucan phosphatase
activity is highly conserved in nature and essential for both starch
and glycogen metabolism.
SEX4 and LSF2 proteins both contain a chloroplast targeting
peptide, a DSP domain, and a unique C-terminal (CT) motif (Kerk
et al., 2006; Niittylä et al., 2006; Sokolov et al., 2006; Gentry et al.,
2007; Kötting et al., 2009; Vander Kooi et al., 2010; Santelia et al.,
2011). Chloroplast targeting peptides localize proteins to the
chloroplast, the site of starch metabolism. The CT motif was
originally identified in the SEX4 structure as a motif that folds into
the DSP core and is essential for protein stability and function
(Vander Kooi et al., 2010). Additionally, SEX4 contains a carbohydrate binding module (CBM) that is common in starch-interacting
enzymes (Niittylä et al., 2006; Sokolov et al., 2006; Gentry et al.,
2007; Glaring et al., 2011). CBMs are nonenzymatic domains that
typically bind a specific carbohydrate and allow the catalytic
portion of the enzyme to modify the substrate (Coutinho and
Henrissat, 1999; Boraston et al., 2004; Machovic and Janecek,
2006). The previously determined glucan-free SEX4 structure
demonstrated that its CBM and DSP domains interact to form
a continuous binding pocket that coordinates the dual functions
of glucan binding and dephosphorylation (Vander Kooi et al.,
2010). An additional plant protein called LSF1 also contains
a CBM and DSP domain (Comparot-Moss et al., 2010). While
LSF1 is required for starch degradation, it lacks phosphatase
activity and is not considered a glucan phosphatase (ComparotMoss et al., 2010). Conversely, LSF2 binds and dephosphorylates glucans, but it is notable that this glucan phosphatase
lacks a CBM (Santelia et al., 2011). Indeed, the glucan phosphatase family was first defined as any protein containing both
a DSP and CBM (Gentry et al., 2007). Thus, the physical basis
for LSF2–starch interaction and dephosphorylation is unclear,
although it has been previously suggested that LSF2 may use
a scaffold protein or an unidentified glucan binding interface to
maintain interactions with starch (Comparot-Moss et al., 2010;
Santelia et al., 2011).
Here, we elucidate the functional basis for LSF2 as a specific
glucan phosphatase by determining the x-ray crystal structure
of LSF2 with and without the phospho-glucan products maltohexaose and phosphate. LSF2 possesses a unique DSP active
site that incorporates both a glucan binding platform and
phosphatase catalytic residues. Moreover, we identify two additional secondary binding sites (SBSs) located >20 Å from the
active site that intimately involve the CT motif and are essential
for LSF2 glucan binding and dephosphorylation, providing the
distinct mechanism necessary for LSF2 to function without
a CBM.
RESULTS
Crystal Structure of LSF2 Bound to Maltohexaose
and Phosphate
The structure of the Arabidopsis LSF2 glucan phosphatase (residues 79 to 282, C193S [catalytically inactive]) bound to maltohexaose and phosphate was determined to a resolution of 2.30 Å
using molecular replacement with one molecule in the asymmetric
unit (Figure 1A, Table 1). The LSF2 DSP domain (residues 79 to
244) possesses a characteristic core PTP fold consisting of a
central five-stranded b-sheet region flanked by eight a-helices
(see Supplemental Figure 1 online). The CT motif (residues 245 to
282) consists of a loop region culminating in an a-helix that integrally folds into the DSP domain, a characteristic also found in
the glucan phosphatase SEX4 (Vander Kooi et al., 2010). A search
for structural homologs of the LSF2 DSP domain (residues 79 to
244) identified the DSP domain of Arabidopsis SEX4 (residues 90
to 250, root mean square deviation [RMSD] of 1.1 Å, PDB code
3NME; Vander Kooi et al., 2010) and mouse PTPMT1 (residues
105 to 256, RMSD of 2.2 Å, PDB code 3RGQ; Xiao et al., 2011) as
the structures most similar to LSF2 despite the fact that the LSF2
DSP is only 48 and 17% identical at the amino acid level to the
DSP domain of SEX4 and PTPMT1, respectively.
Maltohexaose is composed of six Glc moieties with a-1,4glycosidic linkages; thus, it is similar to the unwound helices on
the starch granular surface. In the structure, maltohexaose is
bound to the LSF2 active site and two distal sites. The LSF2
active-site region contains a single maltohexaose chain and
phosphate molecule within the catalytic pocket. Multiple conserved DSP active-site motifs converge to form an extended
active-site binding pocket within LSF2 that is ;19Å long and
;9Å deep with 511 Å2 contact area (Figure 1B). These motifs
include a recognition domain from a1 through b1 (83 to 92),
a variable (V-) loop from a3 through a4 (132 to 150), a WPD (D-)
loop between b4 and a5 (158 to 163), a PTP-loop between b5
and a6 (192 to 199) that contains the LSF2 catalytic signature
(Cx5R) motif, and an R-motif between a7 and a8 (225 to 244)
(see Supplemental Figures 1 and 2 online).
Structure of the Glucan Phosphatase LSF2
3 of 13
Figure 1. Structure of LSF2 Bound to Maltohexaose and Phosphate.
(A) Ribbon diagram of LSF2 (residues 79 to 282, C193S). Maltohexaose chains (green, cyan, orange, and pink) and phosphate (teal) are shown.
Elements of secondary structure are numbered consecutively from N to C termini.
(B) Maltohexaose chain (green) and phosphate (teal) at the active-site (red) channel. Image correlates with the red box in (A). Glc moieties are numbered
from the nonreducing to the reducing end. The total contact area of the active-site channel with phosphate and maltohexaose is 511 Å2.
(C) Specific activity of LSF2 wild type (WT) and inactive mutant LSF2 C193S (C/S) against the C6- and C3-position of Arabidopsis starch. Phosphatefree starch was purified from gwd-deficient Arabidopsis (Yu et al., 2001), and the starch was prelabeled with 33P at either the C6- or C3-position and
then incubated with recombinant protein. Starch dephosphorylation over time was linear and was measured via release of 33P. The reaction time was
5 min. Each bar is the mean + SD of six replicates.
(D) 2Fo-Fc electron density map of maltohexaose chain (green) and phosphate (teal) at the active site (1.35 s). The O3 group of Glc3 is highlighted.
(E) LSF2 active-site catalytic triad (yellow) residues interact with maltohexaose and phosphate. S(C)193 (catalytically inactive mutant, C193S) and R199
are located within the PTP-loop between b5 and a6, and Asp-161 is located within the D-loop between b4 and a5.
LSF2 C3 Specificity and Catalytic State
LSF2 possesses robust activity against starch and displays high
specificity for the C3 position, as measured by a 33P-radiolabeled starch dephosphorylation assay (Figure 1C) (Santelia
et al., 2011). While both glucan dikinases and glucan phosphatases display strong positional specificity, the basis of this
specificity is unclear. The electron density of the maltohexaose
in the active site allows clear assignment of glucan chain position and orientation, labeled Glc1-Glc6 from the nonreducing
end to the reducing (Figure 1D; see Supplemental Figures 3A
and 3B online). Strikingly, the O3 group of Glc3 directly interacts
with the phosphate at the LSF2 catalytic site at a distance of
2.4 Å compared with 7.0 Å for the O6 group. Furthermore, the
orientation of the PTP catalytic triad (DX30-35CX5R) within LSF2
is proximal to the O3 and phosphate and poised for catalysis of
a C3-phosphorylated Glc (Figure 1E). C193S is located at the
base of the active-site cleft, 2.5 Å from the nearest phosphate
oxygen, and represents the key nucleophilic catalytic residue
that covalently attacks the phosphate group during catalysis.
Arg-199 is positioned 2.8 Å from the phosphate and orients the
phosphate of the substrate toward the catalytic Cys. At the
top of the active-site cleft is the D-loop Asp-161 that participates in catalysis as a general acid/base, forming the reaction
intermediate and then assisting in hydrolysis and product expulsion. Asp-161 is located 2.5 Å from the O3 of Glc3 and 3.8 Å
from the nearest phosphate oxygen.
Maltohexaose-Phosphate Product Bound at the LSF2
Active Site
Five highly conserved aromatic residues delineate the boundaries of the extended active-site channel, forming extensive
interactions with the Glc rings of the maltohexaose chain
(Figure 2A). These five aromatic residues provide the majority
of the interface between the LSF2 active site and the substrate. Tyr-83, Tyr-85, Tyr-135, and Trp-136 form one side of
the channel and interact with Glc moieties Glc1-5. Tyr-83 and
Tyr-85 are located within the recognition domain, directly
adjacent to the R-motif and PTP-loop, respectively. Tyr-135
and Trp-136 are both located in the V-loop and form a continuous interaction surface with Tyr-85. Phe-162, located in
the D-loop, forms the opposite side of the aromatic channel
and interacts with Glc2-6. These Glc moieties form a helical
structure around Phe-162 and interact with both faces of
the Phe ring. The residues that form the aromatic channel
are strictly conserved in all land plants as well as in most
single-celled members of Kingdom Plantae, with only one
4 of 13
The Plant Cell
Table 1. Crystallographic Statistics
Crystal
PDB code
Data collection
Space group
Cell dimensions
a, b, c (Å)
a, b, g (°)
Wavelength (Å)
Resolution range
(Å) (highest
shell resolution)
Rmerge (highest shell)
I /sI (highest shell)
Completeness (%)
(highest shell)
Redundancy
(highest shell)
Refinement
Resolution (Å)
No. reflections
Rwork/Rfree
No. atoms
Protein
Ligand/ion
Water
B-factors
Protein
Ligand/ion
Water
RMSDs
Bond lengths (Å)
Bond angles (°)
Ramachandran plot
Most favored
regions (%)
Additional allowed
regions (%)
Disallowed regions (%)
D78-LSF2 C193S:
Maltohexaose,
Phosphate
D78-LSF2 Wild
Type: Citrate
4KYR
4KYQ
P6522
P21212
92,78, 92.78, 144.94
90, 90, 120
1.00
20.0-2.30 (2.38-2.30)
51.82, 99.58, 37.76
90, 90, 90
1.00
20.0-1.65 (1.71-1.65)
10.8 (39.1)
9.0 (2.2)
94.8 (87.3)
6.6 (59.3)
20.5 (2.3)
90.1 (86.3)
3.7 (2.5)
4.7 (4.7)
2.30
15288
17.4/22.9
1.65
20833
15.7/19.9
1685
251
88
1711
13
161
45.4
37.6 (phosphate)
70.1 (maltohexaose,
active site)
59.1 (maltohexaose,
Site-2)
60.8 (maltohexaose,
Site-3, hex-1)
86.1 (maltohexaose,
Site-3, hex-2)
49.7
19.8
14.6 (citrate)
0.014
2.1
0.014
1.7
97.6
97.7
2.4
2.3
0.0
0.0
31.5
nonconservative substitution (Volvox carteri L83; see Supplemental
Figure 2 online).
To investigate the functionality of these aromatic residues, we
generated Ala mutants of each channel residue and tested their
ability to dephosphorylate starch granules isolated from Arabidopsis. Single Ala point mutations of Tyr-83, Tyr-85, Tyr-135, Trp136, and Phe-162 resulted in a decrease of C3-dephosphorylation
by 38 to 96% (Figure 2B). Mutation of both sides of the channel
(W136A/F162A) resulted in a 99% loss of glucan phosphatase
activity. Importantly, the observed decreases in specific glucan
phosphatase activity were not due to destabilization of the active
site or misfolding of the protein as evidenced by near-wild-type
phosphatase activity for all mutant proteins toward the generic
substrate para-nitrophenyl phosphate (pNPP) (see Supplemental
Figure 4A online). Thus, our findings indicate that LSF2 possesses
an aromatic channel that forms an extended active site uniquely
suited to bind to polyglucan substrates and necessary for glucan
phosphatase activity.
Conformational Changes in the LSF2 Active Site
Intriguingly, one of these key aromatic residues, Phe-162, is
found in the D-loop directly after Asp-161, which serves as the
general acid/base as discussed above. In fact, Phe-162 makes
the most contact of any active-site residue with the substrate
glucan (136 Å2, 27% of the total contact area). To compare the
catalytic site of glucan-bound and unbound LSF2, we crystallized LSF2 without maltohexaose and determined the structure
to a resolution of 1.65 Å using molecular replacement (see
Supplemental Figure 5 online; Table 1). The glucan-free LSF2
produced a different crystal form and contained one molecule in
the asymmetric unit and bound citrate from the crystallization
buffer (see Supplemental Figure 5 online). The RMSD of the
glucan and citrate-bound DSP domain (residues 79 to 244)
structures was 1.0 Å. A comparison of product-bound and citratebound LSF2 structures revealed a substrate-dependent rearrangement of the D-loop architecture upon glucan binding
(Figure 2C). In the product-bound structure, the orientations of
Asp-161 and Phe-162 are significantly different. The D-loop aromatic residue Phe-162, important for the specific activity of LSF2,
interacts with multiple Glc moieties of the glucan chain and shifts
toward the V-loop. This movement is associated with a significant
reorientation of the critical general acid/base, residue Asp-161.
Comparison of the two structures reveals that the terminal carboxylate of Asp-161 is 3.1 Å closer to the catalytic Cys and directly in contact with the O3 group of Glc3. To further analyze the
position of Asp-161, we compared the LSF2 DSP with structures
of the glucan phosphatase SEX4 and the prototypical protein
phosphatases VHR (PDB 1VHR; Yuvaniyama et al., 1996) and
SSH-2 (PDB 2NT2; Jung et al., 2007). Analysis of the catalytic
triad from each of these structures reveals that the substratebound orientation of LSF2 Asp-161 is in a catalytically competent
orientation only in the product-bound structure (see Supplemental
Figure 6 online). Thus, LSF2 undergoes a substrate-induced conformational change with the LSF2 product-bound form ideally positioned for catalysis of an O3 phosphorylated glucan substrate.
Most DSPs possess a short-chain hydrophilic reside, S/T/N/
H, at the +1 residue from the general acid/base Asp (Vander
Kooi et al., 2010). However, Phe-162 is invariant in LSF2, and
the corresponding residue is also strictly conserved in SEX4,
Phe-167. We previously demonstrated that mutating SEX4
Phe-167 to a short-chain hydrophilic residue (F167S) resulted
in a 50% decrease in the glucan/pNPP phosphatase activity of
SEX4 F167S (Vander Kooi et al., 2010). Cumulatively, these data
demonstrate an important role for D-loop movement in order
to correctly position the catalytic triad and maximize glucan
phosphatase activity.
Structure of the Glucan Phosphatase LSF2
5 of 13
Figure 2. The LSF2 Aromatic Channel.
(A) A maltohexaose chain (green) interacts with aromatic channel residues at the LSF2 active site. Tyr-83 and Tyr-85 are located in the recognition
domain N terminus from a1. Tyr-135 and Trp-136 are located in the V-loop in a3. Phe-162 is located in the D-loop between b4 and a5. The total contact
area of the aromatic channel residues with maltohexaose is 266 Å2.
(B) Specific activity of aromatic channel mutants against the C3-position of Arabidopsis starch granules. Phosphate-free starch from gwd-deficient
plants (Yu et al., 2001) was purified and prelabeled at the C3- or C6-position with 33P as in Figure 1C. The labeled starch was then incubated with LSF2
wild type (WT) or LSF2 aromatic channel mutants, and starch dephosphorylation was measured via release of 33P. The reaction time was 5 min. Each
bar is the mean + SD of six replicates. Mutated residues are marked with an asterisk in Supplemental Figure 2 online.
(C) Substrate-dependent positional rearrangement of the D-loop. D-loop residues Phe-162 (aromatic channel) and Asp-161 (catalytic residue) lie
between b4 and a5 and undergo significant movement in the maltohexaose/phosphate bound (blue/yellow) versus the unbound (gray/blue) LSF2
structures.
Noncatalytic Glucan Binding Sites
Glucan phosphatases were originally defined as enzymes that
contain both a DSP and CBM domain, and the CBM has been
shown to be critical for endogenous substrate binding and biological activity (Gentry et al., 2007; Gentry and Pace, 2009).
Because LSF2 lacks a CBM, the nature of its substrate binding
ability has been unclear. Therefore, we investigated LSF2 glucan
binding to amylopectin. LSF2 was incubated with amylopectin,
and the amylopectin was then pelleted by ultracentrifugation.
Proteins in the pellet and supernatant were separated by SDSPAGE and visualized by immunoblot analysis. The prototypical
protein phosphatase VHR does not bind amylopectin and was
found in the supernatant, whereas the prototypical glucan
phosphatase SEX4 possesses robust glucan binding and was
largely in the pellet (Figure 3A). LSF2 also robustly binds amylopectin and similar to SEX4 was largely in the pellet. Next, we
sought to define how mutations in the aromatic channel affect
LSF2 glucan binding. The LSF2 W136A/F162A mutant, which
had a 99% decrease in specific glucan phosphatase activity,
showed only a moderate (32%) decrease in amylopectin binding
(Figure 3A). This suggests that while the active site is necessary
for glucan phosphatase activity, other regions primarily determine substrate binding. These data are consistent with LSF2
containing additional glucan binding sites distinct from the activesite aromatic channel.
Indeed, the maltohexaose-bound LSF2 structure revealed two
additional glucan binding sites >20 Å from the active site (Figure
3B). Thus, we hypothesized that one or both of these additional
glucan binding sites could functionally replace a CBM domain
and be critical for the biological activity of LSF2.
One maltohexaose chain is located in a binding pocket (Site2) formed by residues from the DSP domain and CT motif on the
opposite side of the V-loop ;21 Å from the active site. The
maltohexaose chain makes extensive contacts (391 Å2) with
residues in b4 and a5 of the DSP as well as the C terminus of the
CT domain (Figure 4A; see Supplemental Figures 3C and 3D
online). The maltohexaose chain wraps around the CT-loop,
forming hydrogen bonds with Arg-153 and Arg-157 and van der
Waals contact with Trp-180 and Met-155. As with the active-site
residues, Site-2 residues are highly conserved in LSF2 orthologs
(see Supplemental Figure 2 online). To determine the effect of
6 of 13
The Plant Cell
starch granules. Ala mutations of Glu-268, Lys-245, and Phe261 resulted in decreases of specific glucan phosphatase activity of 35 to 87% (Figure 5B). All mutant proteins maintained
near-wild-type pNPP activity, again indicating that the observed
effects are specific (see Supplemental Figure 4C online). As with
Site-2, we also investigated the effect of Site-3 on substrate
binding. We found that the Site-3 mutant F261A, which showed
Figure 3. LSF2 Glucan Binding Sites.
(A) Results from cosedimentation assay of protein and amylopectin
(amylopectin binding assay). Recombinant His-tagged proteins were
incubated with 5 mg/mL amylopectin, amylopectin was pelleted by ultracentrifugation, and proteins in the pellet (P) and supernatant (S) were
separated by SDS-PAGE and visualized by immunoblot analysis. Amylopectin-bound proteins are found in the pellet and unbound proteins are
found in the supernatant.
(B) Surface model of LSF2 showing DSP domain (blue) and CT motif
(green). Maltohexaose chains at the active site (green), Site-2 (cyan), and
Site-3 (orange and pink) are shown.
Site-2 glucan binding on LSF2 activity, we tested the ability of
Ala point mutants as well as a C-terminal truncation to dephosphorylate starch granules. Ala mutations of Trp-180, Met155, Arg-153, and Arg-157 resulted in decreases of specific
glucan phosphatase activity of 24 to 50% (Figure 4B). Truncation of the three C-terminal residues (R280/G281/T282, DRGT)
decreased activity by 46%. All mutant proteins maintained nearwild-type pNPP activity, indicating that the observed effects are
specific (see Supplemental Figure 4B online). While mutation of
Site-2 resulted in a substantial decrease in glucan phosphatase
activity, we also investigated the effect of Site-2 mutants on
substrate binding. We tested the ability of LSF2 R157A, which
showed the greatest reduction in specific activity, to bind amylopectin and found that it displayed a substantial (64%) decrease in amylopectin binding (Figure 4C). This decrease was
markedly greater than that observed for the W136A/F162A
active-site mutant (Figure 3A). These data demonstrate that
Site-2 functions as a glucan binding interface and that this
binding site is important for the biological activity of starch dephosphorylation by LSF2.
Two additional maltohexaose chains were found in a binding
pocket (Site-3) formed by the CT-loop region ;23 Å from
the active site. Five Glc moieties from two maltohexaose
chains (Hex-1 and Hex-2) could be resolved (Figure 5A; see
Supplemental Figures 3E and 3F online). The two chains form
a helical structure, reminiscent of an amylopectin helix, with
a contact area of 338 Å2. LSF2 primarily interacts with Hex-1,
forming hydrogen bonding interactions with Lys-245 and Glu268 and van der Waal’s interactions with Phe-261. As with the
active-site and Site-2 residues, Site-3 residues are highly conserved in LSF2 orthologs (see Supplemental Figure 2 online). To
determine the effect of Site-3 glucan binding on LSF2 activity,
we tested the ability of Ala point mutants to dephosphorylate
Figure 4. LSF2 Glucan Binding Site-2.
(A) A transparent surface model of LSF2 Site-2 showing DSP domain
(blue) and CT motif (green) interaction with the maltohexaose chain
(cyan). Arg-153 and Met-155 are located on b4. Arg-157 is located between b4 and a5. Trp-180 is located on a5. Thr-282 is located on the C
terminus after a11. Glc moieties are numbered from nonreducing to reducing end. The total contact area of Site-2 with maltohexaose is 391 Å2.
(B) Specific activity of Site-2 mutants against the C3-position of Arabidopsis starch granules. Phosphate-free starch from gwd-deficient plants
(Yu et al., 2001) was purified and prelabeled at the C3- or C6-position
with 33P as in Figure 1C. The labeled starch was then incubated with
LSF2 wild type (WT) or LSF2 Site-2 mutants, and starch dephosphorylation
was measured via release of 33P. In addition to single point mutants, starch
dephosphorylation was also determined for LSF2 lacking the C-terminal
residues Arg-280, Gly-281, and Thr-282 (DRGT). The reaction time was
5 min. Each bar is the mean + SD of six replicates. Mutated residues are
marked with a circle in Supplemental Figure 2 online.
(C) An amylopectin binding assay of Site-2 mutant R157A was performed in
a similar manner as described in Figure 3A. LSF2 R157A was cosedimented
with amylopectin and found in the pellet (P) and supernatant (S).
Structure of the Glucan Phosphatase LSF2
7 of 13
5C). Only a mutant combining all three sites, F162A/W136A/
R157A/F261A, resulted in a protein that possessed no detectable glucan binding ability (Figure 5C). Thus, each of the three
glucan binding sites contributes to substrate binding, and none
of the sites are individually sufficient for wild-type levels of
glucan binding. In addition, all of the LSF2 residues directly involved in noncatalytic glucan binding are not conserved in the
CBM-containing SEX4 (see Supplemental Figure 7 online). Taken
together, these data demonstrate that rather than requiring a scaffold protein or CBM, LSF2 possesses three glucan binding sites in
its DSP domain that are each critical for its ability to bind glucans
and function as a specific glucan phosphatase.
DISCUSSION
Figure 5. LSF2 Glucan Binding Site-3.
(A) A transparent surface model of LSF2 at the CT motif loop (green)
showing specific interactions with maltohexaose chains Hex-1 (orange)
and Hex-2 (pink). Lys-245 is located between a8 and a9. Phe-261 is
located between a10 and a11. Glu-268 is located in a11. The total
contact area of Site-3 with the maltohexaose chains is 338 Å2.
(B) Specific activity of Site-3 mutants against the C3-position of Arabidopsis starch granules. Phosphate-free starch from gwd-deficient plants
(Yu et al., 2001) was purified and prelabeled at the C3- or C6-position
with 33P as in Figure 1C. The labeled starch was then incubated with
LSF2 wild type (WT) or LSF2 Site-3 mutants, and starch dephosphorylation
was measured via release of 33P. The reaction time was 5 min. Each bar is
the mean + SD of six replicates. Mutated residues are marked with a triangle
in Supplemental Figure 2 online.
(C) An amylopectin binding assay of Site-3 mutant F261A, Site-2/Site-3
mutant (R157A/F261A), quadruple mutation of active site, Site-2, and
Site-3 (F162A/W136A/R157A/F261A) was performed in a similar manner
as described in Figure 3A. Amylopectin-bound protein is found in the
pellet (P) and unbound protein is found in the supernatant (S).
the greatest reduction in specific activity, showed the most dramatic
(73%) decrease in amylopectin binding (Figure 5C). In comparison
to Site-2, Site-3 mutants showed greater effects on both substrate
binding and specific glucan phosphatase activity.
To determine the contribution of the individual glucan binding
sites to the overall binding of LSF2, we generated combination
mutants. The combination of Site-2 and Site-3 mutations,
R157A/F261A, led to an even more dramatic (87%) decrease in
amylopectin binding, but slight binding still remained (Figure
Phosphatases dephosphorylate each of the four organic macromolecules: proteins, lipids, nucleic acids, and carbohydrates.
While previous studies have defined the structural bases of
phosphatase activity with proteins, lipids, and nucleic acids, the
basis for phosphatase–glucan interaction was previously unknown. Here, we examined the structural and biochemical basis of
glucan phosphatase activity by determining the crystal structure of
LSF2, providing detailed insights into the mechanism of this important class of enzymes.
At the LSF2 active site, an integrated network of aromatic
residues forms an extended binding pocket that allows specific
glucan interaction and dephosphorylation. While these aromatic
residues are within the core DSP domain, in each case they are
uniquely suited to promote phospho-glucan binding. Conservation of aromatic residues in the active site between glucan
phosphatases suggests that this is a key general characteristic
differentiating glucan phosphatases from other phosphatases. It
is particularly notable that both LSF2 and the previously determined SEX4 structure possess an a-helical V-loop containing
conserved aromatic residues, Tyr-135 and Trp-136 in the LSF2
V-loop (Vander Kooi et al., 2010). These residues are integral for
LSF2 glucan binding and dephosphorylation and form the basis of
the conserved theme of an aromatic channel within the active site
of glucan phosphatases. By contrast, the V-loop of PTPs and
DSPs are historically defined by their lack of secondary structure
and highly variable length (Alonso et al., 2003). Phospho-Tyr
phosphatases contain longer V-loops when compared with phospho-Ser/Thr phosphatases in order to generate a deeper binding
pocket that accommodates the longer phospho-Tyr. However, in
each case, the V-loop is indeed a loop (Alonso et al., 2003; Tonks,
2006). Thus, the glucan phosphatase structured V-loop appears to
be a defining characteristic of this phosphatase class.
Aromatic-Glc stacking interactions are a central structural
element for CBMs (Boraston et al., 2004; Machovic and Janecek,
2006) and are found in the catalytic channel of glycosyl hydrolases (Robert et al., 2005; Koropatkin and Smith, 2010). However,
a similar glucan-aromatic channel interface had not been previously observed in any phosphatase. Our crystallography data
revealed that the LSF2 active site maintains interactions with all
six Glc moieties on the maltohexaose chain, thus indicating that
the active site of LSF2 combines both a DSP active site and
glucan binding platform.
8 of 13
The Plant Cell
Indeed, we identified a link between LSF2 substrate binding
and catalysis mediated by residue Phe-162. In addition to being
a part of the active-site aromatic channel, Phe-162 is located at
the +1 position from the catalytic triad residue Asp-161. Rotation of these residues upon glucan binding is required for the
correct catalytic positioning of Asp-161. This connection suggests an inherent mechanism for phosphate recognition that is
tied directly to the architecture of the glucan phosphatase aromatic channel. It is important to note that this may be a general
feature of this enzyme class since this residue is conserved as
an aromatic/long-chain hydrophobic residue in all known glucan
phosphatases, whereas other phosphatases typically possess
short-chain hydrophilic residues at this position (Vander Kooi
et al., 2010).
Despite similarities with other glucan phosphatases, LSF2 is
in fact unique among known glucan phosphatases in that the
enzyme functions independent of a CBM. There has been debate as to whether or not other functionality or possibly bridging
proteins are required for the glucan phosphatase activity of
LSF2 (Comparot-Moss et al., 2010; Santelia et al., 2011). However,
our data clearly establish that LSF2 uses three glucan binding sites
located in the phosphatase domain for carbohydrate binding. The
central function attributed to CBMs is substrate localization, and
our data demonstrate that the noncatalytic glucan binding sites
identified in the LSF2 structural data adopt this functionality. Mutations of Site-2 and Site-3, the two binding sites located away
from the active site, result in dramatic decreases in LSF2 glucan
binding and dephosphorylation, similar to decreases observed for
CBM mutants of SEX4 and laforin (Ganesh et al., 2004; Wang and
Roach, 2004; Gentry et al., 2007). It should be noted that Site-2
and Site-3 both use residues from the glucan phosphatase–
specific CT motif (Vander Kooi et al., 2010; Santelia et al., 2011).
Thus, this unique elaboration on the core LSF2 phosphatase
domain provides novel functionality.
Carbohydrate active enzymes (CAZymes) as defined by the
CAZy database (http://www.cazy.org) are a diverse collection of
enzymes that synthesize and degrade an extremely heterogeneous group of complex carbohydrates and glycoconjugates
(Cantarel et al., 2009). These enzymes cover >250 protein families, including glycoside hydrolases, glycosyltransferases, polysaccharide lyases, carbohydrate esterases, and nonenzymatic
proteins that contain a CBM (Cantarel et al., 2009). A CBM is
a contiguous amino acid sequence with a conserved tertiary fold
that possesses carbohydrate binding ability and is contained
within a carbohydrate-modifying enzyme (Coutinho and Henrissat,
1999; Boraston et al., 2004; Machovic and Janecek, 2006;
Cantarel et al., 2009). Many of the enzymes that synthesize and
degrade carbohydrates utilize a CBM to bind their carbohydrate
substrate and then enzymatically act on the carbohydrate via
a distinct catalytic module. This model of a binding domain and
enzymatic domain is true for the other identified glucan phosphatases (Worby et al., 2006; Gentry et al., 2007; Kötting et al.,
2009; Vander Kooi et al., 2010). Indeed, the glucan phosphatases were originally defined as any protein containing a phosphatase domain and a CBM (Gentry et al., 2007). While LSF2 is
a carbohydrate-modifying enzyme that binds carbohydrates, it
does not contain a classical CBM and is not classified under the
CAZy classification.
Alternatively, LSF2 uses a glucan binding architecture referred
to as secondary binding sites (SBSs) (Robert et al., 2005;
Bozonnet et al., 2007; Cuyvers et al., 2012). SBSs are an
emerging theme found in some glycoside hydrolases (Cuyvers
et al., 2012). Many glycoside hydrolases possess one or more
CBM, but recent structural studies have identified a subset of
glycoside hydrolases that contain both a CBM and SBSs, such
as SusG (Koropatkin and Smith, 2010), or that only possess
SBSs, such as barley (Hordeum vulgare) a-amylase (Kadziola
et al., 1998; Robert et al., 2005), human salivary and pancreatic
a-amylase (Payan and Qian, 2003; Ragunath et al., 2008), and
yeast glucoamylase (Sevcík et al., 2006). Indeed, the two SBSs of
barley a-amylase, which are remote from its glucan binding active
site, are directly involved in substrate binding and hydrolysis and
these two sites act synergistically (Nielsen et al., 2009).
Our results establish that LSF2 independently binds and
dephosphorylates starch. As starch is a complex, insoluble substrate, the presence of multiple glucan binding interfaces may
permit LSF2 to uniquely engage the complex multivalent glucan
Figure 6. Proposed Model of Reversible Phosphorylation at the Starch
Granular Surface during Breakdown (Modified from Streb and Zeeman,
2012).
At night, starch is phosphorylated (red circles) by glucan dikinases,
leading to solubilization of the outer starch granules via unwinding of
amylopectin helices (gray bars). Amylopectin is then hydrolyzed by
amylases into maltose and malto-oligosaccharides. However, these
amylases cannot completely degrade phosphorylated starch. Therefore,
glucan phosphatases must remove phosphate from partially degraded
amylopectin chains. The outer starch surface is then fully degraded by
amylases into malto-oligosaccharides, and the cycle is reset so that the
next starch layer can be phosphorylated and degraded. The inset box
illustrates a model of LSF2-glucan interaction. Amylopectin helices are
composed of two a-1,4-glycosidic–linked chains that are partially unwound at their nonreducing ends due to phosphorylation. LSF2 interacts
with starch via multiple binding sites: The LSF2 active site (highlighted
in red) interacts with the C3-phosphorylated Glc moiety through coordination of the aromatic channel with six Glc units, Site-2 also interacts
with six Glc moieties via hydrogen bonding and van der Waals contacts
at an interface between the LSF2 DSP domain (blue) and the CT motif
(green), and Site-3 interacts with two helical-like glucan chains through
interactions with the CT motif.
Structure of the Glucan Phosphatase LSF2
surface of its endogenous substrate. Indeed, the combined-site
mutants (R157A/F261A and F162A/W136A/R157A/F261A) show
additive effects, implying that the sites function together. This
suggests a model whereby starch binding involves the engagement of longer or multiple glucan chains by the three glucan binding sites on LSF2 (Figure 6). Moreover, the helical glucan
chains at Site-3 are reminiscent of amylopectin, suggesting LSF2
may interact with complex starch granules with distinct helical
characteristics. The functional significance of SBSs in glycosyl
hydrolases has been extensively reviewed and various additional
functions have been postulated, including substrate disruption,
allosteric regulation, enhancing processivity, and relaying of reaction products (Cuyvers et al., 2012). Due to the limitations of
using the short-chain glucan maltohexaose as a ligand, ongoing
studies using more diverse glucan chains will be required to examine possible cooperativity or connectivity between the additional LSF2 glucan binding sites. This may also provide insights
into the position of LSF2 binding and C3-phosphorylation relative
to amylopectin branch points.
These data also give a clearer picture of the role of glucan
phosphatases in the cyclical starch degradation process. The
glucan dikinases phosphorylate the outer Glc moieties of starch
at the C6- and C3-positions, resulting in amylopectin helix
unwinding and local solubilization. These events allow a- and
b-amylases as well as isoamylases to access the outer glucans
and release maltose and malto-oligosaccharides. However,
b-amylase activity is inhibited when a phosphate group is
reached. The glucan phosphatases bind and dephosphorylate
the phospho-glucans to allow further amylolytic activity and
a resetting of the starch degradation cycle (Figure 6). The coordinated phosphorylation of starch Glc at the C6- and C3-positions
is the central signaling event orchestrating starch breakdown
(Blennow and Engelsen, 2010; Kötting et al., 2010; Streb and
Zeeman, 2012). The recent characterization of SEX4 and LSF2
illustrates that plants use a two-enzyme system for phosphate
removal, permitting complete starch catabolism. Our structures
reveal the basis for the glucan binding and C3-specificity of LSF2.
Determination of additional glucan phosphatase structures will
reveal further insights into the mechanism of activity and specificity
of glucan phosphatases. In particular, it will be interesting to determine if the presence of a CBM domain in other family members
not only contributes to substrate binding but also to positionspecific function.
As the central regulatory event governing starch breakdown,
modulation of reversible phosphorylation has the potential to
increase starch yields and produce starches with novel physicochemical properties, thus enhancing manufacturing of feedstock for both food and nonfood applications (Santelia and
Zeeman, 2011). Starch is used in countless applications, many
of which require chemical and physical modifications to improve
and diversify its functionality (Blennow et al., 2002; Santelia and
Zeeman, 2011). Phosphorylation is the only known natural
modification of starch, and highly phosphorylated starches have
many attractive characteristics, including increased hydration
status (Muhrbeck and Eliasson, 1991), decreased crystallinity
(Muhrbeck and Eliasson, 1991), and improved freeze-thaw
stability, viscosity, and transparency (Blennow et al., 2001).
Arabidopsis plants lacking LSF2 activity display increased C3-
9 of 13
phosphorylated starch without adverse effects on plant growth
(Santelia et al., 2011). Genetic manipulation of LSF2 could therefore represent a new means to produce starch with higher phosphate content, particularly in cereal crops containing virtually no
covalently bound phosphate (Blennow et al., 2000). The data
presented here can be used to inform the engineering of glucan
phosphatases to alter their functionality for industrial starch processing, particularly in light of disparities between C6- and C3contributions to starch superstructure. It has been established that
C3-phosphorylation results in greater local hydration of starch
granules compared with C6-phosphorylation; therefore, manipulation of LSF2 represents a potential avenue for manipulation of
overall starch structure (Blennow and Engelsen, 2010). Structural
insights into the unique mechanism of LSF2 are fundamental to
our understanding of starch catabolism and ultimately harnessing
starch as a biomolecule with diverse applications.
METHODS
Cloning, Expression, and Purification of Recombinant Proteins
Primers used in this work are listed in Supplemental Table 1 online.
Cloning of full-length Arabidopsis thaliana LSF2 (AT3G10940) from cDNA
was previously described (Santelia et al., 2011). Based on data from
secondary structure predictions, disorder predictions, sequence similarity
to SEX4, and analysis of LSF2 orthologs, we generated an Arabidopsis
LSF2 construct lacking the first 78 amino acids (D78-LSF2). D78 LSF2
does not contain the chloroplast targeting peptide (predicted to be
residues 1 to 65) along with residues up to the DSP recognition domain.
D78-LSF2 was subcloned into pET28 (Novagen) using NdeI and XhoI sites
to encode a His6 tag, a thrombin cleavage site, and Δ78-LSF2 (Santelia
et al., 2011). We generated an active-site Cys-to-Ser mutation (C193S), an
established technique that generates a catalytically inactive construct
used to trap the substrate and as a negative control in enzymatic assays
(Zhou et al., 1994; Jia et al., 1995; Begley et al., 2006). All point mutants
were generated using a site-directed mutagenesis kit (Agilent) or mutagenesis services (GenScript). Cloning and purification of Arabidopsis
SEX4 lacking the first 89 amino acids (D89-SEX4) (Vander Kooi et al.,
2010) and Hs-VHR (Yuvaniyama et al., 1996) was performed as previously
described. All DNA sequencing (ACGT) was confirmed using MacVector.
Amino acid sequences of LSF2 orthologs were aligned with ClustalW in
MacVector. Expression and purification of D78-LSF2 was performed
similarly to the previously described method for D89-SEX4 (Vander Kooi
et al., 2010). Briefly, BL21-CodonPlus Escherichia coli cells were transformed with expression vectors for the production of native LSF2 protein.
Cells were grown at 37°C in 23 YT media to OD600 = 0.6 to 0.8, placed on
ice for 20 min, induced with 1 mM isopropyl b-D-thiogalactoside, grown at
16°C for 16 h, and harvested by centrifugation. Cells were lysed in 20 mM
Tris-HCl, pH 7.5, 100 mM NaCl, and 2mM DTT and centrifuged, and the
proteins were purified via a Profinia IMAC Ni2+ column (Bio-Rad) with
a Profinia protein purification system (Bio-Rad). Protein was dialyzed in
20 mM Tris-HCl, pH 7.5, 100 mM NaCl, and 2mM DTT overnight at 4°C in
the presence of thrombin. Affinity purified protein was then reverse purified
over the Profinia IMAC Ni2+ column, and the flow-through fraction was
collected. Protein was further purified to homogeneity using a HiLoad 26/60
Superdex 200 size exclusion column (GE Healthcare). Protein used for
enzyme and binding assays was stored in 10% glycerol as a cryoprotectant
and flash frozen for later use.
Recombinant potato (Solanum tuberosum) GWD and recombinant
Arabidopsis PWD for 33P labeling of Arabidopsis starch were purified as
previously described with the following modifications (Ritte et al., 2002;
Kötting et al., 2005). GWD and PWD were transformed into BL21-
10 of 13
The Plant Cell
CodonPlus E. coli cells and expressed similarly to D78-LSF2 as stated
above. GWD was lysed in buffer (50 mM Tris/HCl, pH 7.5, 2.5 mM EDTA,
2.5 mM DTT, and 0.5 mM PMSF), and proteins were purified using an
anion exchange column (Q-sepharose-FF; GE-Healthcare) with a salt
gradient (50 mM Tris/HCl, pH 7.5, 2.5 mM EDTA, 2.5 mM DTT, 0.5 mM
PMSF, and 0.5 NaCl) to elute the protein. Fractions were collected and
protein was further purified using a HiLoad 26/60 Superdex 200 size
exclusion column (GE Healthcare) in new buffer (100 mM MOPS/KOH, pH
7.6, 1 mM EDTA, 2 mM DTT, 0.5 mM PMSF, and 150 mM NaCl). GWD was
then put over a desalting column using a Bio-Scale Mini Bio-Gel P6 desalting
column (Bio-Rad) using a Profinia protein purification system (Bio-Rad). PWD
was lysed in buffer (50 mM HEPES/NaOH, pH 8.0, 300 mM NaCl, 10 mM
imidazole, and 0.5 mM PMSF) and centrifuged, and proteins were purified
using a Profinia IMAC Ni2+ column (Bio-Rad) with a Profinia protein purification system (Bio-Rad). PWD was further purified using a HiLoad 26/60
Superdex 200 size exclusion column (GE Healthcare).
Crystal Structure Determination and Refinement
For glucan-bound crystals, single, high-quality crystals with one molecule
in the asymmetric unit were obtained via hanging drop vapor diffusion
using a Mosquito liquid handling robot (TTPLabtech) using a 200-nL drop
with a 1:3 ratio of D78-LSF2 C193S (4.8 mg/mL) preincubated with 25 mM
maltohexaose (Sigma-Aldrich): 0.1 M diammonium hydrogen phosphate,
pH 5.7, 17% 2-propoanol, and 31% polyethylene glycol 4000 at 18°C.
Single, high-quality D78-LSF2 wild-type crystals, with one molecule in the
asymmetric unit, were obtained with a 200-nL drop using a 1:1 ratio of
LSF2 (4.8 mg/mL): 0.1 M trisodium citrate, pH 5.8, 16% 2-propanol, 31%
polyethylene glycol 4000, and 2% glycerol at 18°C. A single crystal was
used for data collection and structural determination for both LSF2
structures. Both D78-LSF2 C193S (phosphate/maltohexaose) and D78LSF2 wild-type (citrate) data were collected on the 22-ID beamline of SERCAT at the Advanced Photon Source, Argonne National Laboratory (Table
1) at 110K at a wavelength of 1.0 Å. Data were processed using HKL2000
(Otwinowski and Minor, 1997). PHENIX (Adams et al., 2010) was used for
molecular replacement using the SEX4 DSP and a10 helix of the SEX4 CT
domain as search models (Vander Kooi et al., 2010). The structures were
then fully built and refined via iterative model building and refinement
using Coot (Emsley et al., 2010) and Refmac5 (Murshudov et al., 1997),
respectively. Stereochemistry of the model was analyzed using MolProbity (Davis et al., 2007). Analysis and molecular graphics were prepared using Pymol (Schrodinger, 2010). Density maps were produced
using the FTT program in CCP4 (Winn et al., 2011). Comparative structure
analyses were performed using the DaliServer (Holm and Rosenström,
2010) and DaliLite (Hasegawa and Holm, 2009). Protein-ligand contact
analyses were performed with Areaimol (Lee and Richards, 1971).
Phosphorylation with unlabeled ATP at the C3-position by PWD was
performed as previously described (Hejazi et al., 2010). C3-33P-labeled
starch was generated by isolating phosphate-free starch granules from
the Arabidopsis sex1-3 mutant (Yu et al., 2001), phosphorylating the
starch with unlabeled ATP at the C6-position by GWD followed by
phosphorylation with 33P at the C3-position by PWD and washing until all
unincorporated 33P had been removed, as previously described (Hejazi
et al., 2010). In both cases, the starch granules were phosphorylated at
both positions; however, the 33P-label was located at only one or the other
position. [b-33P]ATP was obtained from Hartmann Analytic. Recombinant
proteins (150 ng) were incubated in dephosphorylation buffer (100 mM
sodium acetate, 50 mM bis-Tris, 50 mM Tris-HCl, pH 6.5, 0.05% [v/v] Triton
X-100, 1 µg/mL [w/v] BSA, and 2 mM DTT) with C6- or C3-prelabeled starch
(4 mg/mL) in a final volume of 150 mL on a rotating wheel for 5 min at 25°C.
The reaction was terminated by the addition of 50 mL of 10% SDS. The
reaction tubes were then centrifuged at 13,000 rpm for 5 min to pellet the
starch. 33P release into 150 mL of supernatant was determined using a 1900
TR liquid scintillation counter (Packard). The assay was performed with each
protein six times or more to determine specific activity.
Glucan Binding Assay
Glucan binding assays were performed as previously described with the
following modifications (Gentry et al., 2007; Dukhande et al., 2011). All
proteins used in the glucan binding assays were purified as described above
without cleavage of the His6 tag to maintain the epitope for immunoblot
analysis. Amylopectin, from potato starch (Sigma-Aldrich) was solubilized by
the Roach method (Wang and Roach, 2004) at a concentration of 5 mg/mL.
Five milligrams of amylopectin was then prepelleted via centrifugation at
110,000g for 1.5 h at 4°C to collect only pelletable amylopectin and then
resuspended in 0.5 mL binding buffer (50 mM Tris, 150 mM NaCl, pH 7.5, and
2 mM DTT). One microgram of recombinant protein was incubated in amylopectin solution for an hour at 4°C with rocking. The solution was then
centrifuged at 110,000g for 1.5 h. Cosedimentation with amylopectin was
measured by centrifuging the samples at 110,000g for 1.5 h. All supernatant
was removed and protein was precipitated with 4 volumes of acetone stored
at -20°C. Precipitated protein was then pelleted via centrifugation at 21,130g
for 30 min, and excess acetone was removed using a SpeedVac concentrator (Savant) at 65°C for 1.25 h. Both soluble and pellet fractions were then
resuspended in 30 mL RIPA buffer before the addition of 30 mL SDS-PAGE
buffer (60 mL total volume for both the soluble and pellet fractions). Fifteen
microliters of the pellet and soluble fraction were then resolved via SDSPAGE, and relative concentration of pellet and soluble protein was analyzed
by immunoblotting with a-His6 antibody. Quantification of the signal was
determined using ImageJ (Abramoff et al., 2004). The assay was performed
three times or more with each protein to determine binding capacity.
Phosphatase Assays
Accession Numbers
Phosphatase assays using pNPP have been previously described and
were performed with the following modifications (Worby et al., 2006;
Gentry et al., 2007; Sherwood et al., 2013). Hydrolysis of pNPP was
performed in 50 mL reactions, containing 13 phosphatase buffer (0.1 M
sodium acetate, 0.05 M bis-Tris, 0.05 M Tris-HCl, pH 7.0, and 2 mM DTT),
50 mM pNPP, and 1 µg of enzyme at 37°C for 15 min. The reaction was
terminated by the addition of 200 mL of 0.25 NaOH, and absorbance was
measured at 410 nm. The assay was performed with each protein six
times or more to determine specific activity.
Phosphate release from 33P-Labeled granules was performed as
previously described with the following variations (Hejazi et al., 2010;
Santelia et al., 2011). C6-33P-labeled starch was generated by isolating
phosphate-free starch granules from the Arabidopsis sex1-3 mutant (Yu
et al., 2001), phosphorylating the starch with 33P at the C6-position by
GWD followed by washing until all unincorporated 33P had been removed.
Sequence data from this article can be found in the Arabidopsis Genome
Initiative or GenBank/EMBL databases under the following accession
numbers: LSF2, At3g10940; SEX4, At3g52180; GWD, At1g10760; PWD,
At5g26570; and VHR, Hs03174. The atomic coordinates and structure factors
have been deposited in the Protein Data Bank (PDB code 4KYR and 4KYQ,
for maltohexaose/phosphate and citrate bound structures, respectively).
Supplemental Data
The following materials are available in the online version of this article.
Supplemental Figure 1. DSP Subdomains in LSF2.
Supplemental Figure 2. Conservation of LSF2 Orthologs.
Supplemental Figure 3. Electron Density Maps of Ligands in LSF2
Structure.
Structure of the Glucan Phosphatase LSF2
Supplemental Figure 4. Specific Activity of LSF2 and Mutants against
para-Nitrophenyl Phosphate.
Supplemental Figure 5. Stereo View of Citrate at the LSF2 Active
Site.
Supplemental Figure 6. Structural Alignment of the DSP Catalytic
Triad of LSF2.
Supplemental Figure 7. Sequence Conservation of LSF2 and SEX4.
Supplemental Table 1. Sequences of Primers Used in This Work.
ACKNOWLEDGMENTS
We thank Samuel Zeeman and members of the Gentry and Vander Kooi
labs for fruitful discussions as well as Carol Beach and Martin Chow in
the Molecular Basis of Human Disease COBRE Proteomics Core and
Protein Analytical Core, respectively, for technical assistance. This study
was supported in part by an National Science Foundation CAREER Grant
MCB-1252345 (M.S.G.), National Institutes of Health Grants R01NS070899
(M.S.G.) and P20GM103486 (M.S.G. and C.W.V.K.), Kentucky Science and
Energy Foundation Grant KSEF-2268-RDE-014 (M.S.G.), a University of
Kentucky Summer Research and Creativity Fellowship (T.M.B.), University
of Kentucky College of Medicine startup funds (M.S.G.), the ETH-Zürich
(O.K.), and Swiss-South African Joint Research Program Grant IZ LS
X3122916 (D.S. and O.K.). The contents are solely the responsibility of the
authors and do not necessarily represent the official views of the National
Science Foundation, National Institutes of Health, or other funding
agencies.
AUTHOR CONTRIBUTIONS
M.S.G., C.W.V.K., and D.A.M. designed the research. All authors performed
research. C.W.V.K., D.A.M., H-.F.G., M.S.G., and S.H. analyzed data.
D.A.M., C.W.V.K., and M.S.G. wrote the article, and D.A.M. generated the
figures C.W.V.K. and M.S.G. contributed equally to the publication of this
article.
Received April 16, 2013; revised May 31, 2013; accepted June 12, 2013;
published June 28, 2013.
REFERENCES
Abramoff, M.D., Magalhaes, P.J., and Ram, S.J. (2004). Image
processing with ImageJ. Biophotonics International 11: 36–42.
Adams, P.D., et al. (2010). PHENIX: A comprehensive Python-based
system for macromolecular structure solution. Acta Crystallogr. D Biol.
Crystallogr. 66: 213–221.
Alonso, A., Rojas, A., Godzik, A., and Mustelin, T. (2003). The DualSpecific Protein Tyrosine Phosphatase Family. (Berlin: Springer).
Baunsgaard, L., Lütken, H., Mikkelsen, R., Glaring, M.A., Pham,
T.T., and Blennow, A. (2005). A novel isoform of glucan, water
dikinase phosphorylates pre-phosphorylated alpha-glucans and is
involved in starch degradation in Arabidopsis. Plant J. 41: 595–605.
Begley, M.J., Taylor, G.S., Brock, M.A., Ghosh, P., Woods, V.L., and
Dixon, J.E. (2006). Molecular basis for substrate recognition by
MTMR2, a myotubularin family phosphoinositide phosphatase. Proc.
Natl. Acad. Sci. USA 103: 927–932.
Blennow, A., Bay-Smidt, A.M., Olsen, C.E., and Møller, B.L. (2000).
The distribution of covalently bound phosphate in the starch
11 of 13
granule in relation to starch crystallinity. Int. J. Biol. Macromol. 27:
211–218.
Blennow, A., and Engelsen, S.B. (2010). Helix-breaking news:
Fighting crystalline starch energy deposits in the cell. Trends Plant
Sci. 15: 236–240.
Blennow, A., Mette Bay-Smidt, A., and Bauer, R. (2001).
Amylopectin aggregation as a function of starch phosphate content
studied by size exclusion chromatography and on-line refractive
index and light scattering. Int. J. Biol. Macromol. 28: 409–420.
Blennow, A., Nielsen, T.H., Baunsgaard, L., Mikkelsen, R., and
Engelsen, S.B. (2002). Starch phosphorylation: A new front line in
starch research. Trends Plant Sci. 7: 445–450.
Boraston, A.B., Bolam, D.N., Gilbert, H.J., and Davies, G.J. (2004).
Carbohydrate-binding modules: Fine-tuning polysaccharide recognition.
Biochem. J. 382: 769–781.
Bozonnet, S., Jensen, M.T., Nielsen, M.M., Aghajari, N., Jensen,
M.H., Kramhøft, B., Willemoës, M., Tranier, S., Haser, R., and
Svensson, B. (2007). The ‘pair of sugar tongs’ site on the noncatalytic domain C of barley alpha-amylase participates in substrate
binding and activity. FEBS J. 274: 5055–5067.
Buléon, A., Colonna, P., Planchot, V., and Ball, S. (1998). Starch
granules: Structure and biosynthesis. Int. J. Biol. Macromol. 23:
85–112.
Cantarel, B.L., Coutinho, P.M., Rancurel, C., Bernard, T., Lombard,
V., and Henrissat, B. (2009). The Carbohydrate-Active EnZymes
database (CAZy): An expert resource for glycogenomics. Nucleic
Acids Res. 37 (Database issue): D233–D238.
Caspar, T., Lin, T.P., Kakefuda, G., Benbow, L., Preiss, J., and
Somerville, C. (1991). Mutants of Arabidopsis with altered regulation
of starch degradation. Plant Physiol. 95: 1181–1188.
Comparot-Moss, S., et al. (2010). A putative phosphatase, LSF1, is
required for normal starch turnover in Arabidopsis leaves. Plant
Physiol. 152: 685–697.
Coutinho, P.M., and Henrissat, B. (1999). Carbohydrate-active
enzymes: An integrated database approach. Recent Advances in
Carbohydrate Bioengineering, 246: 3–12.
Cuyvers, S., Dornez, E., Delcour, J.A., and Courtin, C.M. (2012).
Occurrence and functional significance of secondary carbohydrate
binding sites in glycoside hydrolases. Crit. Rev. Biotechnol. 32:
93–107.
Davis, I.W., Leaver-Fay, A., Chen, V.B., Block, J.N., Kapral, G.J.,
Wang, X., Murray, L.W., Arendall, W.B., III., Snoeyink, J.,
Richardson, J.S., and Richardson, D.C. (2007). MolProbity: allatom contacts and structure validation for proteins and nucleic
acids. Nucleic Acids Res. 35 (Web Server issue): W375–W383.
Dukhande, V.V., Rogers, D.M., Romá-Mateo, C., Donderis, J.,
Marina, A., Taylor, A.O., Sanz, P., and Gentry, M.S. (2011).
Laforin, a dual specificity phosphatase involved in Lafora disease, is
present mainly as monomeric form with full phosphatase activity.
PLoS ONE 6: e24040.
Edner, C., Li, J., Albrecht, T., Mahlow, S., Hejazi, M., Hussain, H.,
Kaplan, F., Guy, C., Smith, S.M., Steup, M., and Ritte, G. (2007).
Glucan, water dikinase activity stimulates breakdown of starch
granules by plastidial beta-amylases. Plant Physiol. 145: 17–28.
Emsley, P., Lohkamp, B., Scott, W.G., and Cowtan, K. (2010).
Features and development of Coot. Acta Crystallogr. D Biol. Crystallogr.
66: 486–501.
Fettke, J., Hejazi, M., Smirnova, J., Höchel, E., Stage, M., and
Steup, M. (2009). Eukaryotic starch degradation: Integration of plastidial
and cytosolic pathways. J. Exp. Bot. 60: 2907–2922.
Gallant, D.J., Bouchet, B., and Baldwin, P.M. (1997). Microscopy of
starch: Evidence of a new level of granule organization. Carbohydr.
Polym. 32: 177–191.
12 of 13
The Plant Cell
Ganesh, S., Tsurutani, N., Suzuki, T., Hoshii, Y., Ishihara, T., DelgadoEscueta, A.V., and Yamakawa, K. (2004). The carbohydrate-binding
domain of Lafora disease protein targets Lafora polyglucosan bodies.
Biochem. Biophys. Res. Commun. 313: 1101–1109.
Gentry, M.S., Dixon, J.E., and Worby, C.A. (2009). Lafora disease:
Insights into neurodegeneration from plant metabolism. Trends
Biochem. Sci. 34: 628–639.
Gentry, M.S., Dowen, R.H., III., Worby, C.A., Mattoo, S., Ecker,
J.R., and Dixon, J.E. (2007). The phosphatase laforin crosses
evolutionary boundaries and links carbohydrate metabolism to neuronal
disease. J. Cell Biol. 178: 477–488.
Gentry, M.S., and Pace, R.M. (2009). Conservation of the glucan
phosphatase laforin is linked to rates of molecular evolution and the
glucan metabolism of the organism. BMC Evol. Biol. 9: 138.
Glaring, M.A., Baumann, M.J., Abou Hachem, M., Nakai, H., Nakai,
N., Santelia, D., Sigurskjold, B.W., Zeeman, S.C., Blennow, A.,
and Svensson, B. (2011). Starch-binding domains in the CBM45
family—Low-affinity domains from glucan, water dikinase and a-amylase
involved in plastidial starch metabolism. FEBS J. 278: 1175–1185.
Hansen, P.I., Spraul, M., Dvortsak, P., Larsen, F.H., Blennow, A.,
Motawia, M.S., and Engelsen, S.B. (2009). Starch phosphorylation—
Maltosidic restrains upon 39- and 69-phosphorylation investigated by
chemical synthesis, molecular dynamics and NMR spectroscopy.
Biopolymers 91: 179–193.
Hasegawa, H., and Holm, L. (2009). Advances and pitfalls of protein
structural alignment. Curr. Opin. Struct. Biol. 19: 341–348.
Hejazi, M., Fettke, J., Kötting, O., Zeeman, S.C., and Steup, M.
(2010). The Laforin-like dual-specificity phosphatase SEX4 from
Arabidopsis hydrolyzes both C6- and C3-phosphate esters introduced by
starch-related dikinases and thereby affects phase transition of alphaglucans. Plant Physiol. 152: 711–722.
Holm, L., and Rosenström, P. (2010). Dali server: conservation
mapping in 3D. Nucleic Acids Res. 38 (Web Server issue): W545–
W549.
Jia, Z., Barford, D., Flint, A.J., and Tonks, N.K. (1995). Structural
basis for phosphotyrosine peptide recognition by protein tyrosine
phosphatase 1B. Science 268: 1754–1758.
Jung, S.K., Jeong, D.G., Yoon, T.S., Kim, J.H., Ryu, S.E., and Kim,
S.J. (2007). Crystal structure of human slingshot phosphatase 2.
Proteins 68: 408–412.
Kadziola, A., Søgaard, M., Svensson, B., and Haser, R. (1998). Molecular
structure of a barley alpha-amylase-inhibitor complex: Implications for
starch binding and catalysis. J. Mol. Biol. 278: 205–217.
Keeling, P.L., and Myers, A.M. (2010). Biochemistry and genetics of
starch synthesis. Annu. Rev. Food Sci. Technol. 1: 271–303.
Kerk, D., Conley, T.R., Rodriguez, F.A., Tran, H.T., Nimick, M.,
Muench, D.G., and Moorhead, G.B. (2006). A chloroplast-localized
dual-specificity protein phosphatase in Arabidopsis contains a
phylogenetically dispersed and ancient carbohydrate-binding domain,
which binds the polysaccharide starch. Plant J. 46: 400–413.
Koropatkin, N.M., and Smith, T.J. (2010). SusG: A unique cellmembrane-associated alpha-amylase from a prominent human gut
symbiont targets complex starch molecules. Structure 18: 200–215.
Kötting, O., Kossmann, J., Zeeman, S.C., and Lloyd, J.R. (2010).
Regulation of starch metabolism: The age of enlightenment? Curr.
Opin. Plant Biol. 13: 321–329.
Kötting, O., Pusch, K., Tiessen, A., Geigenberger, P., Steup, M.,
and Ritte, G. (2005). Identification of a novel enzyme required
for starch metabolism in Arabidopsis leaves. The phosphoglucan, water
dikinase. Plant Physiol. 137: 242–252.
Kötting, O., Santelia, D., Edner, C., Eicke, S., Marthaler, T., Gentry,
M.S., Comparot-Moss, S., Chen, J., Smith, A.M., Steup, M.,
Ritte, G., and Zeeman, S.C. (2009). STARCH-EXCESS4 is a laforin-
like Phosphoglucan phosphatase required for starch degradation in
Arabidopsis thaliana. Plant Cell 21: 334–346.
Lee, B., and Richards, F.M. (1971). The interpretation of protein
structures: Estimation of static accessibility. J. Mol. Biol. 55:
379–400.
Machovic, M., and Janecek, S. (2006). Starch-binding domains in the
post-genome era. Cell. Mol. Life Sci. 63: 2710–2724.
Minassian, B.A., et al. (1998). Mutations in a gene encoding a novel
protein tyrosine phosphatase cause progressive myoclonus epilepsy.
Nat. Genet. 20: 171–174.
Moorhead, G.B., De Wever, V., Templeton, G., and Kerk, D. (2009).
Evolution of protein phosphatases in plants and animals. Biochem.
J. 417: 401–409.
Muhrbeck, P., and Eliasson, A. (1991). Influence on the naturally
occuring phosphate esters on the crystallinity of potato starch. J.
Sci. Food Agric. 55: 13–18.
Murshudov, G.N., Vagin, A.A., and Dodson, E.J. (1997). Refinement
of macromolecular structures by the maximum-likelihood method.
Acta Crystallogr. D Biol. Crystallogr. 53: 240–255.
Nielsen, M.M., Bozonnet, S., Seo, E.S., Mótyán, J.A., Andersen,
J.M., Dilokpimol, A., Abou Hachem, M., Gyémánt, G., Naested,
H., Kandra, L., Sigurskjold, B.W., and Svensson, B. (2009). Two
secondary carbohydrate binding sites on the surface of barley
alpha-amylase 1 have distinct functions and display synergy in hydrolysis
of starch granules. Biochemistry 48: 7686–7697.
Niittylä, T., Comparot-Moss, S., Lue, W.L., Messerli, G., Trevisan,
M., Seymour, M.D., Gatehouse, J.A., Villadsen, D., Smith, S.M.,
Chen, J., Zeeman, S.C., and Smith, A.M. (2006). Similar protein
phosphatases control starch metabolism in plants and glycogen
metabolism in mammals. J. Biol. Chem. 281: 11815–11818.
Otwinowski, Z., and Minor, W. (1997). Processing of X-ray diffraction
data collected in oscillation mode. Method. Enzymol. 276: 307–326.
Payan, F., and Qian, M. (2003). Crystal structure of the pig pancreatic
alpha-amylase complexed with malto-oligosaccharides. J. Protein
Chem. 22: 275–284.
Ragunath, C., Manuel, S.G., Venkataraman, V., Sait, H.B.,
Kasinathan, C., and Ramasubbu, N. (2008). Probing the role of
aromatic residues at the secondary saccharide-binding sites of human
salivary alpha-amylase in substrate hydrolysis and bacterial binding.
J. Mol. Biol. 384: 1232–1248.
Ritte, G., Heydenreich, M., Mahlow, S., Haebel, S., Kötting, O., and
Steup, M. (2006). Phosphorylation of C6- and C3-positions of glucosyl
residues in starch is catalysed by distinct dikinases. FEBS Lett. 580:
4872–4876.
Ritte, G., Lloyd, J.R., Eckermann, N., Rottmann, A., Kossmann, J.,
and Steup, M. (2002). The starch-related R1 protein is an alphaglucan, water dikinase. Proc. Natl. Acad. Sci. USA 99: 7166–7171.
Roach, P.J. (2002). Glycogen and its metabolism. Curr. Mol. Med. 2:
101–120.
Robert, X., Haser, R., Mori, H., Svensson, B., and Aghajari, N.
(2005). Oligosaccharide binding to barley alpha-amylase 1. J. Biol.
Chem. 280: 32968–32978.
Santelia, D., Kötting, O., Seung, D., Schubert, M., Thalmann, M.,
Bischof, S., Meekins, D.A., Lutz, A., Patron, N., Gentry, M.S., Allain,
F.H., and Zeeman, S.C. (2011). The phosphoglucan phosphatase like
sex Four2 dephosphorylates starch at the C3-position in Arabidopsis.
Plant Cell 23: 4096–4111.
Santelia, D., and Zeeman, S.C. (2011). Progress in Arabidopsis starch
research and potential biotechnological applications. Curr. Opin. Biotechnol.
22: 271–280.
Schrodinger, LLC. (2010). The PyMOL Molecular Graphics System,
Version 1.3.
Structure of the Glucan Phosphatase LSF2
Serratosa, J.M., et al. (1999). A novel protein tyrosine phosphatase
gene is mutated in progressive myoclonus epilepsy of the Lafora
type (EPM2). Hum. Mol. Genet. 8: 345–352.
Sevcík, J., Hostinová, E., Solovicová, A., Gasperík, J., Dauter, Z.,
and Wilson, K.S. (2006). Structure of the complex of a yeast
glucoamylase with acarbose reveals the presence of a raw starch
binding site on the catalytic domain. FEBS J. 273: 2161–2171.
Sherwood, A.R., Paasch, B.C., Worby, C.A., and Gentry, M.S. (2013). A
malachite green-based assay to assess glucan phosphatase activity.
Anal. Biochem. 435: 54–56.
Sokolov, L.N., Dominguez-Solis, J.R., Allary, A.L., Buchanan, B.B.,
and Luan, S. (2006). A redox-regulated chloroplast protein phosphatase
binds to starch diurnally and functions in its accumulation. Proc. Natl.
Acad. Sci. USA 103: 9732–9737.
Streb, S., and Zeeman, S.C. (2012). Starch metabolism in Arabidopsis.
In The Arabidopsis Book 10: e0160, doi/10.1199/tab.0160.
Tagliabracci, V.S., Turnbull, J., Wang, W., Girard, J.M., Zhao, X.,
Skurat, A.V., Delgado-Escueta, A.V., Minassian, B.A., DepaoliRoach, A.A., and Roach, P.J. (2007). Laforin is a glycogen
phosphatase, deficiency of which leads to elevated phosphorylation
of glycogen in vivo. Proc. Natl. Acad. Sci. USA 104: 19262–19266.
Takeda, Y, and Hizukuri, S. (1981). Studies on starch phosphate.
Part 5. Reexamination of the action of sweet-potato bet-amylase on
phosphorylated (1->4)-a-D-glucan. Carbohydr. Res. 89: 174–178.
Tester, R.F., Karkalas, J., and Qi, X. (2004). Starch - Composition,
fine structure and architechture. J. Cereal Sci. 39: 151–165.
Tonks, N.K. (2006). Protein tyrosine phosphatases: From genes, to
function, to disease. Nat. Rev. Mol. Cell Biol. 7: 833–846.
Tonks, N.K. (2013). Special issue: Protein phosphatases: From molecules
to networks: Introduction. FEBS J. 280: 323.
13 of 13
Vander Kooi, C.W., Taylor, A.O., Pace, R.M., Meekins, D.A., Guo,
H.F., Kim, Y., and Gentry, M.S. (2010). Structural basis for the
glucan phosphatase activity of Starch Excess4. Proc. Natl. Acad.
Sci. USA 107: 15379–15384.
Wang, W., and Roach, P.J. (2004). Glycogen and related polysaccharides
inhibit the laforin dual-specificity protein phosphatase. Biochem. Biophys.
Res. Commun. 325: 726–730.
Winn, M.D., et al. (2011). Overview of the CCP4 suite and current
developments. Acta Crystallogr. D Biol. Crystallogr. 67: 235–242.
Worby, C.A., Gentry, M.S., and Dixon, J.E. (2006). Laforin, a dual
specificity phosphatase that dephosphorylates complex carbohydrates.
J. Biol. Chem. 281: 30412–30418.
Xiao, J., Engel, J.L., Zhang, J., Chen, M.J., Manning, G., and Dixon,
J.E. (2011). Structural and functional analysis of PTPMT1, a phosphatase
required for cardiolipin synthesis. Proc. Natl. Acad. Sci. USA 108: 11860–
11865.
Yu, T.S., et al. (2001). The Arabidopsis sex1 mutant is defective in
the R1 protein, a general regulator of starch degradation in plants,
and not in the chloroplast hexose transporter. Plant Cell 13:
1907–1918.
Yuvaniyama, J., Denu, J.M., Dixon, J.E., and Saper, M.A. (1996).
Crystal structure of the dual specificity protein phosphatase VHR.
Science 272: 1328–1331.
Zeeman, S.C., Tiessen, A., Pilling, E., Kato, K.L., Donald, A.M.,
and Smith, A.M. (2002). Starch synthesis in Arabidopsis. Granule
synthesis, composition, and structure. Plant Physiol. 129: 516–529.
Zhou, G., Denu, J.M., Wu, L., and Dixon, J.E. (1994). The catalytic
role of Cys124 in the dual specificity phosphatase VHR. J. Biol.
Chem. 269: 28084–28090.
Structure of the Arabidopsis Glucan Phosphatase LIKE SEX FOUR2 Reveals a Unique
Mechanism for Starch Dephosphorylation
David A. Meekins, Hou-Fu Guo, Satrio Husodo, Bradley C. Paasch, Travis M. Bridges, Diana Santelia,
Oliver Kötting, Craig W. Vander Kooi and Matthew S. Gentry
Plant Cell; originally published online June 28, 2013;
DOI 10.1105/tpc.113.112706
This information is current as of June 18, 2017
Supplemental Data
/content/suppl/2013/06/24/tpc.113.112706.DC2.html
/content/suppl/2013/06/17/tpc.113.112706.DC1.html
Permissions
https://www.copyright.com/ccc/openurl.do?sid=pd_hw1532298X&issn=1532298X&WT.mc_id=pd_hw1532298X
eTOCs
Sign up for eTOCs at:
http://www.plantcell.org/cgi/alerts/ctmain
CiteTrack Alerts
Sign up for CiteTrack Alerts at:
http://www.plantcell.org/cgi/alerts/ctmain
Subscription Information
Subscription Information for The Plant Cell and Plant Physiology is available at:
http://www.aspb.org/publications/subscriptions.cfm
© American Society of Plant Biologists
ADVANCING THE SCIENCE OF PLANT BIOLOGY